Multi-frame image blood vessel feature identification and information fusion-based color coding method

Abstract

The invention provides a multi-frame image blood vessel feature identification and information fusion-based color coding method. The method comprises the following four steps of 1, reading angiographydata, and preprocessing a single frame image; 2, based on a preprocessing result in the step 1, performing blood vessel feature identification and denoising on the single frame image; 3, performing the operations in the steps 1 and 2 on a multi-frame image sequence, and performing information fusion on multiple frame images; and 4, reading specific data in fused information to perform a color coding algorithm. Firstly a DSA sequence is subjected to enhancement and feature identification, then the processed frame images are subjected to HSV color space coding by applying the specific blood vessel data, and HSV color space is converted into RGB color space for display, so that blood flow information comprised in a whole black-white sequence is displayed in a single color image, and a doctorcan quickly and effectively perform interpretation on an angiography result.

Description

【Technical field】 [0001] The patent of the present invention relates to the field of medical image processing, in particular to a color coding method for multi-frame image vessel feature recognition and information fusion. 【Background technique】 [0002] Cerebrovascular disease is a relatively common disease that seriously affects human quality of life and health, especially among middle-aged and elderly people. Once a stroke occurs, it is very harmful to human beings. The disease is characterized by high disability rate, high mortality rate, and high recurrence rate. It is one of the three major diseases that cause human death. According to the data of the diseases causing death in our country for decades, the hazards of cerebrovascular diseases are very surprising, accounting for the second place in the proportion of deaths. According to the statistical data of our country in recent decades, the average number of people suffering from cerebral apoplexy is about 2 million ...

AI Technical Summary

Problems solved by technology

From the point of view of medical application, among many vascular imaging techniques, CTA and MRA can obtain three-dimensional information of cerebral vascular structure, but the above two types of imaging techniques have defects, they are inaccurate in image locality and details, and doctors cannot find out small details through them. Blood vessels, and estimating the exact location of vascular stenosis and aneurysms

Images